Internet Technology Letters最新文献

Sports nutrition is the balanced diet or diet chart that helps to improve performance of sports persons. It is globally accepted that Indian foods are rich in nutrition. It is greatly preferred by yogis, gurus and dieticians to intake Indian foods in order to maintain a wellness and healthy lifestyle. Smart watches, wearable devices, mobile applications and digital portals are available to suggest foods to the sports persons. But software application based on Indian foods specifically for women athletes are not exists so far. In this paper an intelligent food recommendation system based on Long Short-Term Memory (LSTM) and LSTM with GRU is proposed to suggest meals for women athletes. LSTM has connections and it processes the entire sequence of data thus, it is week suitable for suggestion models. The performance of the model is validated using evaluation metrices and the result demonstrate its effectiveness.

The real-time embedded system is a computer-based special-purpose intelligent computer system that meets the strict requirements of the application system for power consumption. With the development of the Internet era, people are increasingly using the Internet to obtain relevant information, while traditional data collection mainly relies on manual operations. Therefore, the collection, storage and transmission of track and field training information based on existing equipment and databases has become a research hotspot. Based on wireless medical sensor network, this paper proposes an efficient track and field training information collection and feedback system communication protocol for real-time transmission of various data of track and field athletes, and a real-time embedded system for data collection and storage. Through research and analysis, it is found that the collection and feedback of track and field training information based on wireless medical sensor network can improve the accuracy of collection by 8.625. This shows that the collection and feedback of sports training information based on wireless medical sensor network is feasible.

The integration of sensors in cyber-physical systems has given rise to data markets, where data owners can offer their sensing data for sale to potential buyer. However, determining the optimal data price in such markets is a complex issue, which demands a careful consideration of the interests of all parties involved, as well as the potential privacy loss for data sellers. By taking privacy loss into account, this paper proposes a fair compensation model for data sellers and formulates the pricing problem as a Stackelberg game. An automatic data pricing algorithm is developed to calculate the optimal price maximizing the joint benefits of the data sellers and the buyer where the privacy loss of the data sellers are compensated reasonably. Numerical simulations validate the effectiveness of the proposed pricing model in balancing benefits and privacy loss.

Chinese culture is depicted in a profound manner through opera music. With the advancements in deep learning and IoT technology, numerous studies have increasingly utilized neural networks to supersede conventional acoustic models. This paper explores the emotion classification of Qinqiang Opera through the utilization of cutting-edge research methods. Firstly, we improve the convolutional neural network and adopt the residual network model to increase the model's fitting and stability. Secondly, the attention mechanism is integrated to reinforce the expression of each weight information, allowing the network to differentiate feature information more effectively and elevating the overall performance of the network. Thirdly, we use five sensors to form a local Internet of Things to collect a large amount of Qin opera audio data for experiments. Finally, multiple experiments confirm the effectiveness of the proposed model in the emotional classification of Qinqiang Opera.

This paper proposes a physiological signal analysis method in exercise fatigue detection application based on deep learning models to provide fast and accurate feedback for the player's physical status and better assist the player to perform exercise. We adopt the deep neural network as backbone model and design following strategies in our proposed method to process and extract features in signals. First, we preprocess the physiological signal, including noise reduction and segmentation. Second, we use a deep learning model to design a feature extraction method, which uses an autoencoder to label and feature the signal. Third, we perform motion fatigue detection on the fused signal features based on a long short-term memory network model. The results prove that the method proposed has good performance.

Language intelligence analysis oriented to multi-sensor data fusion is of great significance for language analysis in real scenarios. On the one hand, intelligent language analysis technology can greatly improve the performance of applications such as information retrieval and machine translation, and provide technical support for semantic-level applications. On the other hand, each language has its own unique characteristics, and the advancement of the language system through language analysis technology is of great benefit to natural language analysis. In this letter, an intelligent language analysis method for multi-sensor data fusion is elaborated. Specifically, the Kalman filter algorithm is combined to perform the first preprocessing filter fusion on multi-sensor data. Then, the deep learning model is used to design a language analysis model using Bidirectional Long-Short Memory Neural Networks (Bi-LSTM) to obtain deep fusion of multi-sensor data. In the experiment, the multi-sensors are used to collect real language data and public language datasets for verification, and the results show the effectiveness of the method proposed in this letter in terms of syntactic label classification.

Yoga, as a kind of body building exercise, has always been loved by people. However, many people suffer from yoga training injuries due to long-term incorrect posture and wrong exercise methods. There is an urgent need for a technology to help people detect and improve yoga training methods. Based on the past computer assistance method, this paper started from a new idea, and adopted the method of multi-sensor information fusion to detect yoga training, aiming to help the masses better participate in yoga training. In this study, 50 volunteers were invited to participate in the comparative experiment. Based on multi-sensor information fusion, and by building a human model, the tension and compression data before and after yoga training were compared to analyze the differences before and after calculation. It was concluded that the more sensors, the higher the degree of information fusion, and the lower the yoga training injury index. The injury index of yoga training without multi-sensor information fusion technology in the early stage was 0.39. With the increase of the number of sensors, the injury index of yoga training has gradually decreased to 0.02, which was more than 5 percentage points lower than that of the previous methods. The experiment showed that the method of yoga training damage detection based on multi-sensor information fusion was feasible, which also provided a new idea for the research of yoga training injury detection methods.